Gao Report Fed Oversite Hcbs Needs Strengthened

8/14/2019 Gao Report Fed Oversite Hcbs Needs Strengthened

1/70

Report to Congressional RequestersUnited States General Accounting Office

GAO

June 2003 LONG-TERM CARE

Federal Oversight of Growing MedicaidHome andCommunity-BasedWaivers Should BeStrengthened

GAO-03-576


2/70

From 1991 through 2001, Medicaid long-term care spending more thandoubled to over $75 billion, while the proportion spent on institutional caredeclined. Over a similar time period, HCBS waivers grew from 5 percent to19 percent of such expendituresfrom $1.6 billion to $14.4 billionand thenumber of waivers, participants, and average state per capita spending alsogrew significantly. Since 1992, the number of waivers increased by almost70 percent to 263 in June 2002, and the number of beneficiaries, as of 1999,had nearly tripled to almost 700,000, of which 55 percent were elderly.

In the absence of specific federal requirements for HCBS quality assurancesystems, states provide limited information to the Centers for Medicare &Medicaid Services (CMS), the federal agency that administers the Medicaid

program, on how they assure quality of care in their waiver programs for theelderly. States waiver applications and annual reports for waivers for theelderly often contained little or no information on state mechanisms forassuring quality in waivers, thus limiting information available to CMS thatshould be considered before approving or renewing waivers. GAOs analysisof available CMS and state waiver oversight reports for waivers serving theelderly identified oversight weaknesses and quality of care problems. Morethan 70 percent of the waivers for the elderly that GAO revieweddocumented one or more quality-of-care problems. The most common

problems included failure to provide necessary services, weaknesses in plans of care, and inadequate case management. The full extent of such problems is unknown because many state waivers lacked a recent CMSreview, as required, or the annual state waiver report lacked the relevantinformation.

CMS has not developed detailed state guidance on appropriate qualityassurance approaches as part of initial waiver approval. Although CMSoversight has identified some quality problems in waivers, CMS does notadequately monitor state waivers and the quality of beneficiary care. The 10CMS regional offices are responsible for ongoing monitoring for HCBSwaivers. However, CMS does not hold these offices accountable forcompleting periodic waiver reviews, nor does it hold states accountable forsubmitting annual reports on the status of waiver quality. Consequently,CMS is not fully complying with statutory and regulatory requirements whenit renews waivers. As of June 2002, almost one-fifth of waivers in place for 3

years or more had either never been reviewed or were renewed without areview; for an additional 16 percent of waivers, reports detailing the reviewresults were never finalized. Regional office personnel explained thatlimited staff resources and travel funds often impede the timing and scope of reviews. While regional office reviews include record reviews for a sampleof waiver beneficiaries, they do not always include beneficiary interviews.The reviews also varied considerably in the number of beneficiary recordsreviewed and their method of determining the sample.

Home and community-basedsettings have become a growing

part of states Medicaid long-termcare programs, serving as analternative to care in institutionalsettings, such as nursing homes. Tocover such services, however,states often obtain waivers fromcertain federal statutoryrequirements. GAO was asked toreview (1) trends in states use of Medicaid home and community-based service (HCBS) waivers,

particularly for the elderly, (2) statequality assurance approaches,including available data on thequality of care provided to elderlyindividuals through waivers, and(3) the adequacy of federaloversight of state waivers.

GAO is recommending that the Administrator of CMS take steps to(1) better ensure that state qualityassurance efforts are adequate to

protect the health and welfare of HCBS waiver beneficiaries, and(2) strengthen federal oversight of the growing HCBS waiver

programs. Although CMS raisedcertain concerns about aspects of the report, such as the respectivestate and federal roles in qualityassurance and the potential needfor additional federal oversightresources, CMS generallyconcurred with therecommendations.

www.gao.gov/cgi-bin/getrpt? GAO-03-576 .

To view the full product, including the scopeand methodology, click on the link above.For more information, contact Kathryn G.Allen at (202) 512-7118.

Highlights of GAO-03-576 , a report tocongressional requesters

June 2003

LONG-TERM CARE

Federal Oversight of Growing MedicaidHome and Community-Based WaiversShould Be Strengthened
http://www.gao.gov/cgi-bin/getrpt?GAO-03-576http://www.gao.gov/cgi-bin/getrpt?GAO-03-576http://www.gao.gov/cgi-bin/getrpt?GAO-03-576http://www.gao.gov/cgi-bin/getrpt?GAO-03-576http://www.gao.gov/cgi-bin/getrpt?GAO-03-576http://www.gao.gov/cgi-bin/getrpt?GAO-03-576http://www.gao.gov/cgi-bin/getrpt?GAO-03-576


3/70

Page i GAO-03-576 Medicaid Home and Community-Based Waivers

Letter 1

Results in Brief 3Background 5Waivers Are Vehicle for Dramatic Growth in Medicaid Home and

Community-Based Services 10Information on State Quality Assurance Approaches for Waivers

Serving the Elderly Is Limited, but Quality Concerns Have BeenIdentified 14

CMS Guidance to States and Oversight Of HCBS Waivers AreInadequate to Ensure Quality Care 22

Conclusions 34Recommendations for Executive Action 35

Agency and State Comments and Our Evaluation 35

Appendix I Scope and Methodology 42

Appendix II Suggested CMS Definitions of Home andCommunity-Based Services in Waivers Servingthe Elderly 45

Appendix III Medicaid Long-Term Care Expenditures, by Typeand State, Fiscal Year 2001 47

Appendix IV Number of Beneficiaries Served by HCBS Waivers for the Elderly and in Nursing Homes,by State, 1999 49

Appendix V Number of HCBS Waivers for the Elderly,Beneficiaries, Expenditures, and per BeneficiaryExpenditures by State, 1999 51

Contents


4/70

Page ii GAO-03-576 Medicaid Home and Community-Based Waivers

Appendix VI CMS HCBS Quality Initiatives 53

Appendix VII Beneficiary Samples for and Duration of Regional Office Reviews of 15 State WaiversServing the Elderly 56

Appendix VIII Comments from the Centers for Medicare &Medicaid Services 58

Tables

Table 1: States with Highest and Lowest per BeneficiaryExpenditures for State HCBS Waivers Serving the Elderly,1999 13

Table 2: Quality Assurance Mechanisms States Reported Using inHCBS Waivers Serving the Elderly 15

Table 3: Quality Assurance Mechanisms Frequently Cited in Waiver

Applications and Current Annual State Reports for HCBSWaivers Serving the Elderly 17Table 4: Frequently Cited Quality-of-Care Problems Identified by

CMS Regional Offices or States in HCBS Waivers Servingthe Elderly 21

Table 5: HCBS Waivers That Had 10 Years or More Elapse withoutEver Having a Regional Office Review or without a ReviewPrior to the Last Waiver Renewal, as of June 2002 25

Table 6: Status of CMS and State Monitoring for the 15 LargestHCBS Waivers Serving the Elderly 28

Table 7: Number and Specialty of CMS Regional Office Staff Assigned to Oversee HCBS Waivers

Table 8: Services States May Include in Their Medicaid Home andCommunity-Based Services Waiver 45

Figure

Figure 1: Percentage Distribution of Medicaid Long-Term CareExpenditures, Fiscal Years 1991 and 2001 11


5/70

Page iii GAO-03-576 Medicaid Home and Community-Based Waivers

Abbreviations

CMS Centers for Medicare & Medicaid ServicesFTE full-time equivalentHCBS home and community-based servicesHCFA Health Care Financing AdministrationHHS Department of Health and Human ServicesICF/MR intermediate care facility for the mentally retarded

This is a work of the U.S. Government and is not subject to copyright protection in theUnited States. It may be reproduced and distributed in its entirety without furtherpermission from GAO. It may contain copyrighted graphics, images or other materials.Permission from the copyright holder may be necessary should you wish to reproducecopyrighted materials separately from GAOs product.


6/70

Page 1 GAO-03-576 Medicaid Home and Community-Based Waivers

June 20, 2003

The Honorable Charles E. GrassleyChairmanCommittee on FinanceUnited States Senate

The Honorable John B. BreauxRanking Minority MemberSpecial Committee on AgingUnited States Senate

Over the last decade, states have increased their support for long-termcare services in individuals homes or in other community-based settingssuch as adult day care, adult foster care homes, and assisted livingfacilitiesas an alternative to care in nursing homes and otherinstitutions. For many vulnerable elderly and nonelderly individuals with

physical, developmental, or cognitive disabilities, these alternative settingsand services are seen as preferable to institutional care. Most state fundingof long-term care is through Medicaid, the federal-state health care

program for certain low-income individuals. Medicaid home andcommunity-based services (HCBS) waivers, authorized under section1915(c) of the Social Security Act, are the primary means by which states

provide noninstitutional long-term care. 1 Waivers allow states to limit theavailability of services geographically, target specific populations orconditions, control the number of individuals served, and cap overallexpendituresactions not usually allowed under the Medicaid statute.The Centers for Medicare & Medicaid Services (CMS)the federal agencythat manages Medicaidreviews and approves states requests for thesewaivers and also is responsible for ensuring that states have necessarysafeguards to protect the health and welfare of individuals receivingservices through waiver programs. 2

142 U.S.C. 1396n(c)(2000).2Until June 2001, CMS was known as the Health Care Financing Administration (HCFA). Inthis report, we continue to refer to HCFA when our findings apply to the organizationalstructure and operations associated with that name.

United States General Accounting Office Washington, DC 20548


7/70


Despite the growing use of HCBS waivers, concerns have been raisedabout the quality of care provided through waivers serving both elderlyand nonelderly populations. Newspaper exposs and some state auditreports have chronicled serious health and welfare concerns in waiver

programs across the country. Because of continued growth in the numbersof people served through HCBS waiver programs and concerns about thequality of care, you asked us to review (1) trends in states use of suchwaivers, particularly for the elderly, (2) state quality assurance approachesfor waivers serving the elderly, including available data on the quality of care provided to beneficiaries, and (3) the adequacy of CMSs oversight of state waiver programs for the elderly as well as those for other target

populations.

To identify trends in states use of waivers, we analyzed CMS and statereports that contained data on waiver beneficiaries, expenditures, andservices. To identify those waivers that serve the elderly, we compiled alist of HCBS waivers with the aged or aged and disabled as their target

populations. Throughout this report, we refer to this universe of waiversas those serving the elderly. To assess state quality assurance activitiesfor waivers serving the elderly, we analyzed (1) data on quality assuranceapproaches from state waiver applications and their most recent annualreports to CMS, (2) the oversight findings reported by states in theirannual waiver reports, and (3) CMS regional office waiver reviews andstate audits of waivers completed from October 1998 through May 2002. 3 For a more in-depth perspective on states quality assurance approachesfor waivers serving the elderly, we conducted structured interviews withstate officials and staff in South Carolina, Texas, and Washington. Weselected these states because they operate some of the largest HCBSwaivers for the elderly that have been in effect for 5 years or longer. Wedid not attempt to assess the effectiveness of their quality assuranceapproaches. To determine the adequacy of CMS oversight of state waiver

programs for the elderly as well as those for other target populations, weobtained relevant data from officials at CMS headquarters and conductedstructured interviews with all 10 CMS regional offices on their waiver

review activities and staffing as of June 2002. See appendix I for a detaileddiscussion of our scope and methodology. We conducted our review fromNovember 2001 through June 2003 in accordance with generally acceptedgovernment auditing standards.

3Our analysis of regional office waiver reviews is based on final reports. Reviews that didnot have a final report were not included in our analysis.


8/70


Total Medicaid spending for long-term care increased from $33.8 billion infiscal year 1991 to $75.3 billion in fiscal year 2001, with a growing sharespent on services through home and community-based waivers as analternative to care in institutions such as nursing homes. Expenditures forservices through HCBS waivers increased from $1.6 billion in fiscal year1991 to $14.4 billion in fiscal year 2001, growing from 5 percent of allMedicaid long-term care spending in fiscal year 1991 to 19 percent in fiscal

year 2001. Over roughly the same time period, the number of HCBSwaivers increased from 155 to 263, with 77 serving the elderly as of June2002. Every state except Arizona operates at least one waiver for theelderly. From 1992 to 1999, the total number of persons served throughwaivers nationwide nearly tripled to 688,152 and the number of beneficiaries served by waivers for the elderly more than doubled to377,083. In two states, Oregon and Washington, HCBS waiver serviceshave replaced nursing homes as the dominant means of providing long-term care to the elderly under Medicaid. Nationally, average Medicaidexpenditures per beneficiary in waivers serving the elderly increased from$3,622 in 1992 to $5,567 in 1999; average spending per beneficiary in 1999ranged from $1,208 in New York to $15,065 in Hawaii, reflectingdifferences in the type and amount of services provided under differentwaivers.

No nationwide data are available on states quality assurance approachesor the status of quality of care for beneficiaries served by waivers for theelderly, but concerns have been identified about the quality of care

provided under many of these waivers. Because CMS has not provideddetailed guidance to states on federal requirements for HCBS qualityassurance systems, the information available to CMS that should beconsidered before approving or renewing waivers is limited. Thus, statewaiver applications and annual waiver reports that we reviewed forwaivers serving the elderly often contained little or no information onstate quality assurance approaches. For example, 11 applications for the15 largest waivers serving the elderly identified three or fewer specificquality assurance approaches, and none mentioned important approaches

such as complaint systems or enforcement tools. Moreover, 18 of 52 stateannual waiver reports that we reviewed contained no information onapproaches used to help ensure quality. Where information was provided,the most frequently cited quality assurance approaches included (1) auditsor reviews of case management agencies, (2) state agency reviews of waiver providers or direct-care staff, and (3) state licensure, certification,or standards for some waiver providers. Although CMS regional office andstate reviews identified few if any specific cases of harm to waiverbeneficiaries, the reviews for the majority of waivers serving the elderly

Results in Brief


9/70


with available relevant detail had one or more problems related to qualityof care. Among the most commonly cited problems were (1) failure to

provide authorized or necessary services, (2) inadequate assessment ordocumentation of beneficiaries care needs in the plan of care, and(3) inadequate case management. For example, one recent CMS regionaloffice review found that more than one-fourth of a states waiverbeneficiaries had received none of their authorized personal care services.However, the consequences for the beneficiaries were not identified in thisreview. Since many state waiver programs did not have a recent CMSreview, as required, or the annual state waiver report lacked the relevantinformation, the extent of quality-of-care problems is unknown.

CMS guidance to states and oversight of HCBS waivers is inadequate toensure quality of care for waiver beneficiaries. CMS has not developeddetailed guidance for states on appropriate quality assurance mechanismsas part of the waiver approval process, and initiatives under way togenerate information on state quality assurance approaches do notaddress this problem. In addition, the agency has not fully complied withthe statutory and regulatory requirements that condition the renewal of HCBS waivers on (1) states submitting required annual reports thatinclude information on state quality assurance approaches anddeficiencies identified through state monitoring and (2) CMSs conductingand documenting periodic waiver reviews to determine whether statessatisfied requirements for protecting the health and welfare of waiverbeneficiaries. Many state annual waiver reports submitted to CMS regionaloffices for waivers serving the elderly were not timely and lacked requiredinformation on quality assurance and state monitoring. As of June 2002,228 HCBS waivers for all target populations had been in place for 3 yearsor longer and should have been reviewed by CMS regional offices.However, 42 waivers serving approximately 132,000 beneficiaries eitherhad never been reviewed or were renewed without a review. For 36additional waivers, reviews were conducted, but the reports summarizingthe findings were never finalized, raising a question as to whether anyweaknesses were identified and, if so, had been corrected. CMS regional

office personnel informed us that limited staff and travel resourcesimpeded the timing and scope of reviews. While regions reviews includedan examination of beneficiary records, we found that the reviews variedconsiderably in the number of beneficiary records reviewed and theirmethod of determining the sample, raising a question about the extent towhich findings could be generalized. In addition, they did not alwaysinclude beneficiary interviews. Although updated in 2001, CMS guidancefor conducting waiver reviews does not address key operational issues


10/70


such as an adequate sample size or the sampling methodology to provide abasis for generalizing review findings.

To better ensure that state quality assurance efforts are adequate to protect the health and welfare of HCBS waiver beneficiaries and tostrengthen federal oversight, we are recommending that the CMS

Administrator (1) establish more detailed criteria regarding the necessarycomponents of an HCBS waiver quality assurance system, (2) requirestates to submit more specific information about their quality assuranceapproaches prior to waiver approval, (3) ensure that states providesufficient and timely information in their annual waiver reports on theirefforts to monitor quality, (4) develop guidance on the scope andmethodology for federal reviews of state waiver programs, and (5) ensureallocation of sufficient resources for conducting thorough and timelyreviews of quality in HCBS waivers and hold regional offices accountablefor completing such reviews. Although CMS raised certain concerns aboutaspects of our report, such as the respective state and federal roles inquality assurance and the potential need for additional federal oversightresources, the agency generally concurred with our recommendations.

The jointly funded federal-state Medicaid program is the primary source of financing for long-term care services. 4 About one-third of the total$228 billion in Medicaid spending in fiscal year 2001 was for long-termcare in both institutional and community-based settings. States administerthis program within broad federal rules and according to a state planapproved by CMS, the federal agency that oversees and administersMedicaid. Some services, such as nursing home care and home healthcare, are mandatory services that must be covered in any state that

participates in Medicaid. Other services, such as personal care, areoptional, which a state may choose to include in its state Medicaid planbut which then must be offered to all individuals statewide who meet itsMedicaid eligibility criteria. States may also apply to CMS for a section1915(c) waiver to provide home and community-based services as an

alternative to institutional care in a hospital, nursing home, or

4While the purpose of Medicaid is to cover health care and long-term care for low-income persons, including persons who are aged, blind, or disabled, it has become a significantmeans of funding long-term care for many middle-income persons as well. Many of these

persons qualify for Medicaid benefits after a period of spend-down, during which theydeplete their own resources to pay for services.

Background


11/70


intermediate care facility for the mentally retarded (ICF/MR). 5 If approved,HCBS waivers allow states to limit the availability of servicesgeographically, to target services to specific populations or conditions, orto limit the number of persons served, actions not generally allowed forstate plan services. States often operate multiple waivers serving different

population groups, such as the elderly, persons with mental retardation ordevelopmental disabilities, persons with physical disabilities, and childrenwith special care needs.

States determine the types of long-term care services they wish to offerunder an HCBS waiver. Waivers may offer a variety of skilled services toonly a few individuals with a particular condition, such as persons withtraumatic brain injury, or they may offer only a few unskilled services to alarge number of people, such as the aged or disabled. 6 The wide variety of services that may be available under waivers includes home modification,such as installing a wheelchair ramp, transportation, chore services,respite care, nursing services, personal care services, and caregivertraining for family members. CMSs waiver application form for statesincludes a list of home and community-based services with suggesteddefinitions. States are free to include as many or as few of these as theywish, to include additional services, or to include different definitions of services from those supplied with the form. See appendix II for a list of services provided through the HCBS waivers serving the elderly and CMSssuggested definitions of these services.

To be eligible for waiver services, an individual must meet the statescriteria for needing the level of care provided in an institution, such as anursing home, and be able to receive care in the community at a cost

5Federal statutory requirements for Medicaid that may be waived include(1) statewideness, which requires that services be available throughout the state,

(2) comparability, which requires that all services be available to all eligible individuals,and (3) income and resource rules, which require states to use a single income andresource standard when determining eligibility for Medicaid, with the exception of institutional care. A waiver of this last requirement allows states to use more generousinstitutional eligibility criteria when determining financial eligibility for waiver services,thus extending eligibility to individuals in the community who would not otherwise qualifyfor Medicaid.6 A recent summary by the National Association of State Medicaid Directors identified 75discretely defined services in HCBS waiver applications as of June 2000. Individual waiversincluded as few as one service to as many as 25.


12/70


generally not exceeding the cost of institutional care. 7 States areresponsible for determining the specific financial and functional eligibilitycriteria used, conducting the necessary screening and assessment, andarranging for services to be provided. Factors that states use in assessingfunctional eligibility for nursing home care and for waiver services includethe individuals medical condition and their degree of physical or mentalimpairment. Other factors that states generally consider, and which mayaffect the states ability to provide care in the community at a cost notexceeding that of institutional care or to adequately protect beneficiarieshealth and welfare, include the mix of services needed by the individual,the availability of needed services, the cost of services, the need for homemodification, and the availability of family members or other caregivers. 8

In order to receive federal funds for waiver services, a state must submitan application to the Secretary of Health and Human Services (HHS) thatidentifies the target population, specifies the number of persons that willbe served, and lists the services to be included. In addition, states arerequired to provide certain assurances that necessary safeguards havebeen taken to assure financial accountability and to protect the health andwelfare of beneficiaries under the waiver. 9 Federal regulations specify thatthe states safeguards for the health and welfare of beneficiaries mustinclude (1) adequate standards for all providers of waiver services and(2) assurance that any state licensure or certification requirements for

providers of waiver services are met. 10 CMS requires that a states waiverapplication include documentation regarding the standards applicable foreach service provider. If the only requirement for a particular provider is

7The average cost of community care under a waiver cannot exceed the average cost of care in an institution.8For example, a person who requires 24-hour care and supervision and has no family orother support in the community may exceed the limits of what the waiver program allowsin terms of personal care services. However, the same person who lives with a familycaregiver might be eligible to receive several hours of personal care services each day as

well as occasional respite care and caregiver training for the family.9 A state must provide several additional assurances, including the following: (1) the statewill provide for an evaluation of the need for services for individuals, (2) beneficiaries willbe informed of available alternatives to the waiver and provided a choice, (3) the average

per capita expenditures for waiver beneficiaries will not exceed the amount that the stateestimates would have been spent in the absence of the waiver, (4) absent the waiver,beneficiaries would receive the appropriate institutional care that they need, and (5) thestate will provide information to CMS annually on the impact of the waiver.10 See, 42 CFR 441.302(a).


13/70


licensure or certification, the state must provide a citation to theapplicable state statute or regulation. If other requirements apply, the statemust specify the applicable standards that providers must meet andexplain how the provider standards will ensure beneficiaries welfare.Finally, states must annually report on, among other things, how theyimplement, monitor, and enforce their health and welfare standards andthe waivers impact on the health and welfare of beneficiaries.

Initial waiver applications and amendments to initial waivers are reviewedand approved by CMS headquarters. CMSs 10 regional offices have

primary responsibility for reviewing and approving applications to renewwaivers and amendments to renewed waivers. If CMS determines that awaiver application meets program requirements, including sufficientdocumentation to indicate that necessary safeguards are in place to

protect the health and welfare of waiver beneficiaries, it will approve aninitial waiver for a 3-year period. Subsequently, waivers may be extendedfor additional 5-year periods.

Section 1915(c)(3) of the Social Security Act provides that, upon requestof a state, HCBS waivers may be extended, unless the Secretary of HHSdetermines that the assurances provided during the preceding term havenot been met. 11 Among the assurances that the state makes are thatnecessary safeguards have been taken to protect the health and welfare of waiver participants and that the state will submit annual reports on theimpact of the waiver on the type and amount of medical assistance

provided under the state Medicaid plan and on the health and welfare of recipients. Regulations implementing section 1915(c) provide that anextension of a waiver will be granted unless (1) CMSs review of the priorwaiver period shows that the assurances the state made were not met and(2) the state fails to provide adequate documentation and assurances to

justify an extension. 12 In its explanation of this regulation, HCFA indicatedthat a review of the prior period is an indispensable part of the renewal

process. 13

1142 U.S.C. 1396n(c)(3). Section 1915(c)(3) states "A waiver under this subsection [1915(c)]shall be for an initial term of three years and, upon the request of a State, shall be extendedfor additional five-year periods unless the Secretary determines that for the previouswaiver period the assurances provided under paragraph (2) have not been met."1242 CFR 441.304(a).13 See , 59 Fed. Reg. 37702, 37712 (1994) and 53 Fed. Reg. 19950 (1988).


14/70


Reviews of waiver programs for which a renewal has been requested are,therefore, expected to occur at some point during the initial 3-year period,and at least once during each renewal cycle. CMS guidance on the reviewscalls for on-site visits that include an examination of beneficiary and

provider records as well as interviews with state officials. If a statesefforts to protect the health and welfare of waiver beneficiaries aredetermined to be inadequate, CMS officials told us that the agency caneither bar the state from enrolling any new waiver beneficiaries untilcorrective actions are taken or terminate the waiver.

According to a recent CMS-sponsored review, oversight of waivers is oftendecentralized and fragmented among a variety of agencies and levels of government, and rarely does a single entity have accountability for theoverall quality of care provided to waiver beneficiaries. 14 Some waiverservice providers are regulated by state licensing agencies, some arecertified by private accreditation organizations, and others operate underterms of a contract or other agreement with a state agency. While the stateMedicaid agency is ultimately accountable to the federal government forcompliance with the requirements of the waivers, it may delegateadministration of the waivers to state units on aging, mental healthdepartments, or other departments or agencies with jurisdiction over aspecific population or service. About one-third of waivers for the elderlyare administered by an agency or department other than the Medicaidagency, most often the state unit on aging. 15 These agencies may thencontract with local networks, agencies, or providers to provide or arrangefor beneficiary services.

14Maureen Booth and others, Literature Review: Quality Management and Improvement Practices for Home and Community-Based Care (Portland, Me.: University of SouthernMaine, Edmund S. Muskie School of Public Service, Jan. 10, 2002).15Data gathered by the National Association of State Medicaid Directors identified thelocation of waiver administration for 56 HCBS waivers for the elderly as of March 18, 2002.Thirty-eight of these were administered either directly by the Medicaid agency or within thesame department that houses the Medicaid agency.


15/70


16/70


Figure 1: Percentage Distribution of Medicaid Long-Term Care Expenditures, FiscalYears 1991 and 2001

Note: GAO analysis of HCFA Form 64 data as reported by Brian Burwell, Steve Eiken, and KateSredl in Medicaid Long Term Care Expenditures in FY 2001 (The MEDSTAT Group, May 10, 2002).The figure includes data from 49 states and the District of Columbia.

Both the number and size of HCBS waivers have grown considerably overthe past 20 years. Every state except Arizona operates at least one suchwaiver for the elderly. 17 In 1982, the first year of the waiver program, 6states operated HCBS waivers. By 1992, 48 states operated a total of 155HCBS waivers. As of June 2002, 49 states and the District of Columbia

operated a total of 263 HCBS waivers, with 77 serving the elderly. Theaverage waiver for the elderly served 3,305 Medicaid beneficiaries in 1992

17 Arizona operates its Medicaid program as a demonstration project under a section 1115waiver, which includes long-term care as well as acute health care services. Under section1115 of the Social Security Act, the Secretary of HHS has broad authority to authorizeexperimental, pilot, or demonstration projects that are likely to promote objectives of certain federal programs, including Medicaid.


17/70


and 5,892 beneficiaries in 1999. 18 In 1999, 15 states served more than 10,000 persons in their waivers for the elderly, an increase from only 4 states in1992.

The total number of HCBS waiver beneficiarieselderly and nonelderlynationwide nearly tripled from 235,580 in 1992 to 688,152 in 1999, the mostrecent year for which data were available. The number of beneficiariesserved in waivers for the elderly more than doubled from 155,349 in 1992to 377,083 in 1999. Over this same period, the number of Medicaidbeneficiaries who used some nursing home care during the year grew byonly 2.5 percent from 1.57 million to 1.61 million beneficiaries. By 1999,waivers for the elderly were serving 19 percent of all Medicaidbeneficiaries served either in a nursing home or through an HCBS waiverfor the elderly, an increase from 9 percent in 1992. 19 In two states, Oregonand Washington, more elderly and disabled Medicaid beneficiaries wereserved in HCBS waivers in 1999 than were served in nursing homes.

Appendix IV includes the number of Medicaid beneficiaries served byHCBS waivers for the elderly and in nursing homes in each state.

In 1999, the average per beneficiary expenditure in HCBS waivers servingthe elderly was $5,567, an increase from $3,622 in 1992. 20 However, theaverage per beneficiary expenditure for such waivers varied widely acrossstates, reflecting differences in the type, number, and amount of services

provided under waivers in different states. As shown in table 1, amongthose states with waivers serving the elderly in 1999, per beneficiaryexpenditures ranged from an average of $15,065 in Hawaii to $1,208 in

18Waiver beneficiary and expenditure data used in this analysis do not cover the same time periods. Waiver expenditure data are available through 2001. Data on waiver beneficiariesand services are available only through 1999. A CMS contractor recently developed adatabase for HCBS waivers. It is scheduled for installation at CMS in 2003, and it willinclude waiver beneficiary, service, and expenditure data from annual state reports.19The shift from institutional care to home and community-based services under Medicaid

has been most significant for persons with mental retardation or developmentaldisabilities. In 1992, 28 percent of such beneficiaries who qualified for institutional carewere served under HCBS waivers, and by 1999, that proportion had grown to 68 percent.20These average expenditures do not include expenditures for nonwaiver Medicaid servicesfor these beneficiaries. In addition to waiver services, waiver beneficiaries are eligible forthe full range of regular Medicaid state plan services. The overall cost to Medicaid forwaiver beneficiaries will be higher than the amounts reported here, which only includethose services provided under the waiver. In addition, Medicaid covers the cost of roomand board for beneficiaries in nursing homes and other institutions, a benefit not generallycovered for those receiving services under the waiver.


18/70


New York. In Hawaii, one such waiver that provided an average of 85hours of personal assistance services per month to 91 percent of beneficiaries of that waiver had an average cost of $10,893 per beneficiary.

A second Hawaii waiver that provided adult foster care, residential care,or assisted living for waiver beneficiaries had an average cost of $16,958

per beneficiary. In contrast, New Yorks waiver for the elderly did notinclude personal care or residential services; the primary benefits includedsocial work services, personal emergency response systems, and home-delivered meals. Appendix V provides summary information on statesHCBS waivers for the elderly, including per beneficiary expenditures.

Table 1: States with Highest and Lowest per Beneficiary Expenditures for StateHCBS Waivers Serving the Elderly, 1999

StateAverage expenditures

per beneficiaryNumber of

beneficiariesUnited States $5,567 377,083States with highest per beneficiary waiver spendingHawaii 15,065 923New Mexico 14,151 1,404North Carolina 13,778 11,159Alaska 12,015 712West Virginia 11,213 3,470States with lowest per beneficiary waiver spendingMichigan 2,632 6,328Iowa 2,517 3,994Missouri 2,224 20,821Massachusetts 1,919 5,132New York 1,208 19,732

Source: CMS.

Notes: GAO analysis of annual state waiver report data (HCFA Form 372) as reported by CharleneHarrington in Medicaid 1915(c) Home and Community-Based Waivers: Program Data, 1992-1999 (San Francisco, Calif.: University of California, San Francisco, August 2001).

All states in this table except Hawaii operated one waiver serving the elderly in 1999. Hawaii operatedtwo waivers, one that served 288 beneficiaries at a cost of $10,893 per beneficiary and a second thatserved 635 beneficiaries at a cost of $16,958 per beneficiary.


19/70


No comprehensive nationwide data are available on states qualityassurance systems for or the quality of care provided through HCBSwaivers, including those serving the elderly. In the absence of detailedfederal requirements for HCBS quality assurance systems, states waiverapplications and annual reports often contained little or no information onthe mechanisms used to ensure quality, raising a question as to whetherCMS had adequate information to approve or renew some waivers. Morethan half of the waivers serving the elderly for which we were able toobtain a CMS waiver oversight report, an annual state waiver report, or astate audit report identified oversight weaknesses and quality-of-care

problems. Frequently cited quality-of-care problems included (1) failure to provide authorized or necessary services, (2) inadequate assessment ordocumentation of beneficiaries care needs in the plan of care, and(3) inadequate case management. We were unable to analyze over one-third of waivers serving the elderly because they lacked a recent regionaloffice review, the annual state waiver report lacked the relevantinformation, or they were too new to have annual state reports.

Although the state waiver applications and annual waiver reports wereviewed for waivers serving the elderly identified more than a dozenquality assurance approaches, many contained little or no informationabout how states ensure quality. 21 For example, 11 applications for the 15largest waivers serving the elderly identified three or fewer qualityassurance mechanisms and none of these 11 waivers mentioned importantapproaches, including complaint systems or sanctions. Eighteen of 52state annual waiver reports that we reviewed contained no information onthe mechanisms used to help ensure quality. Moreover, when waiverapplications and annual waiver reports did contain some information, theinformation was often incomplete. Our work in South Carolina, Texas, andWashington identified additional quality assurance mechanisms that werenot listed in their waiver applications or annual reports, suggesting thatsuch documents may understate the nature and extent of their oversight

21CMS uses the waiver applications, in part, to assess whether the proposed qualityassurance mechanisms are sufficient to warrant waiver approval. HCFA Form 372, referredto in this report as the annual state waiver report, is a key source of information on howstates have ensured quality until states renew their waivers. In addition to service use andspending data, the annual state waiver report includes information about the states

process for monitoring waiver standards and safeguards and the findings of thosemonitoring processesspecifically, any deficiencies that were detected during the periodcovered by the report.

Information on StateQuality Assurance

Approaches forWaivers Serving theElderly Is Limited, butQuality ConcernsHave Been Identified

States Use a Variety of Waiver Quality Assurance

Approaches in Waivers

Serving the Elderly, YetSome States ProvideLimited or IncompleteInformation to CMS


20/70


approaches. As a result, CMSs understanding of how these states ensurequality in the waivers may be incomplete.

Information provided to CMS in state waiver applications and annualreports identified a variety of mechanisms used to protect the health andwelfare of beneficiaries in waivers serving the elderly. Table 2 describes 14quality assurance approaches that states reported using in HCBS waiversfor the elderly. Some of these approaches focus on the waiver beneficiary,such as case management or beneficiary satisfaction surveys. Otherapproaches are focused on providers, including licensure and inspections,corrective action plans, sanctions, and program manuals. States mayrequire that certain providers be licensed or certified or meet otherrequirements contained in state laws or regulations. Such providers aregenerally subject to periodic inspections that may include a review of beneficiary records to determine whether the records meet programstandards. A third set of quality assurance approaches focuses on waiver

program operations, including internal or external evaluations of thewaiver program, supervisory reviews of waiver beneficiary assessmentsand plans of care, and audits or reviews of case management agencies.

Table 2: Quality Assurance Mechanisms States Reported Using in HCBS WaiversServing the Elderly

Quality assurance mechanism DescriptionBeneficiary-oriented mechanismsCase management Case management includes assessing the

beneficiarys needs, developing the plan of care,arranging for the delivery of services, monitoringthe beneficiary, and conducting periodicreassessments of the beneficiarys needs andmodifying the plan of care as needed.

Beneficiary satisfaction surveys orinterviews

A survey instrument or other tool is used tomeasure waiver beneficiaries views about theirwaiver services and the extent to which servicesare meeting their long-term care needs.

On-site visits of beneficiaries On-site visits may be conducted by programofficials other than the beneficiarys case

manager to observe services being providedand gather information about the care provided.Complaint systems Systems to accept, investigate, and track the

status of waiver beneficiaries or otherscomplaints regarding the waiver program.

Provider-oriented mechanismsLicensure, certification, or other statestandards

States require that certain providers belicensed, certified, or meet other requirementscontained in state law or regulation. Providersare generally subject to periodic inspections that

States Use a Variety of Quality Assurance Mechanisms


21/70


Quality assurance mechanism Descriptioninclude a review of beneficiary records todetermine if they meet program standards.

Provider or direct care staff reviews oraudits

State program officials conduct reviews ofwaiver providers or individual caregivers todetermine whether waiver-specific requirementswere met. Such reviews involve reviews ofbeneficiary records and other providerdocumentation as well as individual beneficiaryinterviews.

Corrective action plans List of actions that the provider agrees to take toreturn to compliance with federal or statestandards.

Sanctions and penalties Depending on the severity of the violation,actions available to penalize the provider for notcomplying with federal or state standards.

Training and technical assistance Ongoing, continuing education for casemanagers and waiver providers to ensurecompetency in delivering and monitoring thecare of waiver beneficiaries.

Program manuals Distribution of rules, policies, procedures, orstandards to waiver providers.

Program-oriented mechanismsCase management agency review oraudit

Reviews of agencies responsible for casemanagement of the HCBS waiver, including areview of a sample of case managers recordsto ensure timeliness and completeness.

Supervisory review of beneficiary

assessments or plans of care

Review conducted by case managers

supervisors or at the state level of documentsrelated to waiver beneficiaries assessed needsand identified services.

Analysis of automated waiver programdata

Review or monitoring of electronic version ofclient data, such as assessments,reassessments, and care plans.

Internal or external evaluation ofwaiver program

Program review of the procedures for waiverbeneficiary assessments, development of plansof care, and delivery of waiver services; reviewmay be conducted by state agency officials orby contractor.

Source: CMS.

Note: GAO analysis of the most recent waiver application for the 15 largest HCBS waivers serving theelderly and the most recent annual state reports for 52 waivers serving the elderly submitted to CMSregional offices as of July 2002.

Because CMS has not provided detailed guidance to states on federalrequirements for HCBS quality assurance systems, the waiver applicationsand annual reports submitted by states to CMS for waivers serving theelderly often contained little or no information on state mechanisms forensuring quality, raising a question as to whether CMS had adequateinformation to approve or renew some waivers.

States Provide CMS LimitedInformation about TheirQuality Assurance Approaches


22/70


Waiver applications. Our review of the most current waiver applicationsfor the 15 largest waivers serving the elderly found that many states

provided CMS limited information about how they plan to protect thehealth and welfare of beneficiaries. 22 Eleven of the 15 states cited three orfewer quality assurance mechanisms. For example, New Yorks applicationonly contained information about the state licensure and certificationrequirements for its waiver services. None of these 11 applicationsincluded well-recognized quality assurance tools such as complaintsystems, corrective action plans, sanctions, or beneficiary satisfactionsurveys. The remaining 4 states each identified six to eight qualityassurance approaches, including at least one of these four important tools.

As shown in table 3, the two mechanisms most frequently cited by stateswere (1) licensure for some HCBS waiver providers, such as home healthagencies and residential care providers, and (2) case management.

Table 3: Quality Assurance Mechanisms Frequently Cited in Waiver Applicationsand Current Annual State Reports for HCBS Waivers Serving the Elderly

Quality assurance mechanism

Waiver application:number of states

citing mechanism (n=15largest state waivers

for the elderly)

Annual statereport: numberof states citing

mechanism a

(n=40 states)Case management agency reviews or

audits 8 30Waiver provider or direct-care staffreviews or audits 1 24Licensure, certification, or other statestandards 15 22Waiver beneficiary satisfaction surveysor interviews 2 21Case management 12 20Training and technical assistance 0 20On-site visits of waiver beneficiaries 1 16Complaint systems 1 13Supervisory review of waiver beneficiaryassessments or plans of care 7 11Corrective action plans 2 9

Sanctions and penalties 1 7

22We reviewed waiver applications for the 15 largest state waivers for the elderly based onthe number of beneficiaries. These waivers were from the following states: Colorado,Florida, Georgia, Illinois, Kentucky, Missouri, New York, North Carolina, Ohio, Oregon,South Carolina, Texas, Virginia, Washington, and Wisconsin. In 1999, these waivers rangedin size from 10,514 beneficiaries in Virginia to 27,978 beneficiaries in Texas.


23/70


Quality assurance mechanism

Waiver application:number of states

citing mechanism (n=15largest state waivers

for the elderly)

Annual statereport: numberof states citing

mechanism a

(n=40 states)Analysis of automated waiver programdata

1 4

Internal or external evaluations of waiverprogram

0 4

Waiver program manuals 0 4 Source: CMS.

Note: GAO analysis of the most recent waiver application for the 15 largest HCBS waivers serving theelderly and the most recent annual state reports for 52 waivers serving the elderly submitted to CMSregional offices as of July 2002.aWe reviewed 70 annual state waiver reports from 49 states and the District of Columbia. Fifty-two ofthese annual reports from 40 states contained some information about states monitoring processesfor HCBS waivers serving the elderly. States may have more than one HCBS waiver serving theelderly.

Annual waiver reports. Compared to waiver applications, annual statewaiver reports identified more quality assurance mechanisms for waiversserving the elderly. The quality assurance mechanisms states annualreports cited most frequently included (1) audits of case managementagencies, (2) reviews of provider or direct-care staff, (3) licensure andcertification of providers, (4) beneficiary satisfaction surveys orinterviews, (5) case management, and (6) training and technicalassistance. As shown in table 3, these six mechanisms were mentioned byat least half of the 40 states that provided such information. 23 However, aswas the case with most of the 15 waiver applications we reviewed,complaint systems, corrective action plans, and sanctions were identifiedless frequently. For example, only 13 of the 40 states identified complaintsystems for waivers serving elderly beneficiaries as a monitoring tool intheir annual waiver reports. 24 Responding to beneficiary complaints is akey element in protecting vulnerable nursing home residents and home

23

As of June 2002, there were 77 waivers serving the elderly. However, our analysis includes2 additional waivers for the elderly that had been terminated or not renewed as of that datebecause the states were able to provide us with their most recent annual report.24Only 1 of the 15 waiver applications we reviewed indicated that the state had a complaintsystem for the providers under its waiver. For a discussion of the role of complaintsystems, see U.S. General Accounting Office, Nursing Homes: Sustained Efforts Are

Essential to Realize Potential of the Quality Initiatives , GAO/HEHS-00-197 (Washington,D.C.: Sept. 28, 2000) and U.S. General Accounting Office, Medicare Home Health Agencies:Weaknesses in Federal and State Oversight Mask Potential Quality Issues , GAO-02-382 (Washington, D.C.: July 19, 2002).
http://www.gao.gov/cgi-bin/getrpt?GAO/HEHS-00-197http://www.gao.gov/cgi-bin/getrpt?GAO/HEHS-00-197http://www.gao.gov/cgi-bin/getrpt?GAO/HEHS-00-197http://www.gao.gov/cgi-bin/getrpt?GAO-02-382http://www.gao.gov/cgi-bin/getrpt?GAO-02-382http://www.gao.gov/cgi-bin/getrpt?GAO-02-382http://www.gao.gov/cgi-bin/getrpt?GAO-02-382http://www.gao.gov/cgi-bin/getrpt?GAO/HEHS-00-197


24/70


health beneficiaries. Moreover, 18 of the elderly waiver reports (26 percent) from 12 states did not include a description of the process formonitoring the standards and safeguards under the waiver, as required onthe reporting form.

State officials in South Carolina, Texas, and Washington informed us theyuse a wider range of quality assurance mechanisms in their waiver

programs than were described in either their waiver application or theirannual state waiver report. Officials in Washington informed us they use12 of the 14 mechanisms identified in table 3, yet they included only 2 of these on their application and 3 in their most recent annual report. Forexample, Washington operates a complaint system for waiver providersbut did not refer to this approach in its waiver application or annualreport. On the other hand, only Washington included reviews or audits of case managers or case management agencies in its application or annualreport, yet all three states provided information on their use of this qualityassurance tool during our interviews. States formal reports to CMS ontheir quality assurance mechanisms may therefore understate the natureand extent of their oversight approaches.

Although information on the quality of care provided in the 79 waiver programs serving the elderly is limited, state oversight problems wereidentified by CMS regional offices or states in 15 of 23 waivers and quality-of-care problems in 36 of 51waivers that we were able to examine. 25 Wewere unable to analyze findings related to 28 waivers serving the elderlyfor various reasons: they lacked a current regional office review or awaiver review report was never finalized, 26 the annual state waiver reportlacked the relevant information, or the waivers were too new to have anannual state report. Because of incomplete information and the absence of

25Our analysis of state oversight issues is based on 23 discrete waivers that had either aregional office review or a state audit. State auditors are responsible for reviewing state

programs and may include Medicaid HCBS waiver programs as a part of these audits. Annual state waiver reports do not address state oversight weaknesses. Our analysis of quality-of-care issues is based on 51 discrete waivers that had either a regional officereview or an annual state report. As of June 2002, there were 77 waivers serving the elderly.However, our analysis of state oversight and quality-of-care problems included 2 additionalwaivers for the elderly that had been terminated or not renewed as of that date becausethey had had a regional office review during the October 1998 through May 2002 time

period we examined.26Regional office review reports that did not have a final report were not included in ouranalysis.

State Oversight and

Quality Issues in WaiversServing the Elderly HaveBeen Identified by CMSRegional Offices andStates


25/70


current reviews for many of the active waivers, the extent of quality-of-care problems is unknown.

CMS regional office reviews or state audits identified weaknesses in stateoversight for waivers serving the elderly in 15 of the 23 waivers weexamined. In some cases, the waiver programs did not have essentialoversight systems or processes in place. For example, in the case of a

Virginia assisted living waiver that had over 1,250 beneficiaries, thePhiladelphia regional office found several state oversight problems,including (1) no system in place to track the completion of the requiredannual resident assessments, (2) insufficient monitoring to ensure thatbeneficiaries were cared for in settings able to meet their needs,(3) insufficient monitoring to ensure that state standards were met forbasic facility safety and hygiene, and (4) failure to inspect medicationadministration records sufficiently to ensure that medication was beingdispensed safely and by qualified staff. The regional office identifiedserious lapses in Virginias oversight of the waiver and the protection of beneficiaries, resulting in both medical and physical neglect of waiverbeneficiaries. On the basis of the regional office review findings, HCFAallowed the waiver to expire in March 2000. In other cases, states mayhave had an oversight system or process in place, but they weredetermined to be inadequate. Five state audit agency reports we reviewedidentified inadequate monitoring systems in state waiver programs. Forexample, Connecticut had a policy in place for monitoring and evaluatingits HCBS waiver program, but, from January 2000 through March 2001 itconducted no quality assurance reviews of the agencies it contracted withto coordinate and manage services for waiver beneficiaries.

CMS regional office reviews and states annual waiver reports identifiedquality-of-care related problems in 36 of 51 HCBS waiver programs for theelderly that we were able to examine. Specifically, they found weaknessesin the delivery of key elements of home and community-based servicesthat could affect waiver beneficiaries health and welfare (see table 4).Typically, the reports did not provide sufficient detail to demonstrate the

impact of these weaknesses on waiver beneficiaries. Consequently, few, if any, specific cases of beneficiary harm were identified.

State Oversight Weaknesses

Quality-of-Care RelatedProblems


26/70


Table 4: Frequently Cited Quality-of-Care Problems Identified by CMS Regional Offices or States in HCBS Waivers Serving theElderly

Problem area Example

Number of 51waivers in which

problem was identifiedProvision of authorizedor necessary services

Beneficiary not receiving services identified as being needed. 20

Plan of care Beneficiarys care needs not addressed in plan of care. 20Case management Case manager for HCBS waiver program not providing ongoing

assessment and monitoring of waiver beneficiaries orinadequate follow-up of changes in beneficiaries care needs.

20

Staffing Insufficient number of staff to provide adequate care or staff nothaving appropriate credentials or training to provide care.

12

Assessment Beneficiarys needs not assessed or reassessment notcompleted in a timely manner.

11

Documentation of service delivery Incomplete record of waiver services provided to beneficiary. 8Training Case managers identified as needing additional training on

Medicaid eligibility.8

Quality assurance orquality of care

HCBS waiver program lacked a formal quality assurancesystem; poor quality of care or services were identified.

7

Medication Unable to document that facilities providing care to waiverbeneficiaries dispensed medication safely and by qualified staff.

4

Source: CMS.

Notes: GAO analysis of CMS regional office final waiver review reports for HCBS waivers serving theelderly issued from October 1998 to May 2002 and the most recent annual state waiver reports for 51waivers serving the elderly.Fifteen waivers serving the elderly had no problems identified in their regional office reviews orannual state reports; the remaining 36 waivers had problems related to quality of care. When both theCMS regional office and the state identified a waiver as having the same type of problem, we countedthat problem only once.

The most frequently identified quality-of-care problems in waivers servingthe elderly involved failure to provide authorized or necessary services,inadequate assessment or documentation of beneficiaries care needs inthe plan of care, and inadequate case management.

Provision of authorized or necessary services. Identified problems

included (1) services identified in plans of care not rendered,(2) inadequate nutrition provided to waiver beneficiaries, and(3) discontinuation of services without adequate notice to beneficiaries.For example, CMSs Dallas regional office found that significant numbersof Oklahoma waiver beneficiaries did not receive personal care servicesfrom their direct-care provider4,303 beneficiaries (27 percent) receivednone of their authorized personal care services and 7,773 beneficiaries (49

percent) received only half of their authorized services. While theconsequences for beneficiaries were not identified in this review, failure to


27/70


provide authorized needed services may result in harm and could affectthe continued ability of beneficiaries to be cared for at home.

Plan of care. Issues included plans of care that (1) insufficiently addressedthe needs of waiver beneficiaries, (2) were not completed or updatedappropriately, and (3) were missing from beneficiaries files. In the reviewof one of the Florida waivers, CMSs Atlanta regional office staff foundseveral instances where needs identified through individual assessments,including significant changes in waiver beneficiaries conditions, were notaddressed in the plan of care, a situation that could lead to beneficiariesnot receiving the necessary services. Without an appropriate plan of careto direct the type and amount of services to be delivered, the waiver

beneficiary may not receive an adequate level of care. Case management. Examples of case management problems included casemanagers who (1) were unaware of beneficiaries having lapses in deliveryof care, (2) were not always aware of procedures or protocols forreporting abuse, neglect, or exploitation, (3) failed to complete residentassessmentsservice plans were either incomplete or inappropriate, andupdates to plans of care were late, or (4) did not always appear to have aclear understanding of service definitions or requirements of the waiver orMedicaid program.

CMS has not developed detailed guidance for states on appropriate quality

assurance approaches as part of the initial waiver approval process.Moreover, although CMS oversight has identified some quality problems, itdoes not adequately monitor HCBS waiver programs or the quality of care

provided to waiver beneficiaries for waivers serving the elderly as well asthose serving other target populations. 27 CMS does not hold its regionaloffices accountable for conducting and documenting periodic waiverreviews, nor does CMS hold states accountable for submitting annualreports on the status of quality in their waivers. As of June 2002, aboutone-fifth of the 228 waivers in place for 3 years or more had either neverbeen reviewed or were renewed without a review. 28 We found that thereviews varied considerably in the number of beneficiary records

27Because CMS regional offices have responsibility for oversight of all HCBS waivers,including those serving the elderly, our analysis included all HCBS waivers as of June 2002.28 As of June 2002, CMS regional offices had oversight responsibility for 263 HCBS waivers.These waivers included other population groups as well as those serving the elderly. Of thistotal, 228 had been in place for 3 years or more and should have had a regional officereview; 70 of these 228 waivers served the elderly. Nine waivers serving the elderly had notbeen in place for 3 years or more and therefore were not included in this analysis.

CMS Guidance toStates and OversightOf HCBS Waivers AreInadequate to EnsureQuality Care


28/70


examined and the method of determining the sample, potentially limitingthe generalizability of findings. According to CMS regional office staff, theallocation of staff resources and travel funding levels have at timesimpeded the scope and timing of their reviews. In addition, some regionaloffice staff told us that limited travel funds have resulted in thesubstitution of more limited desk reviews for on-site visits and in theconduct of reviews with one staff member when two would have been

preferable.

CMS has a number of initiatives under way to generate information anddialogue on quality assurance approaches, but the agencys initiatives stopshort of (1) requiring states to submit detailed information on their qualityassurance approaches when applying for a waiver or (2) stipulating thenecessary components for an acceptable quality assurance system. CMSrecognizes that insufficient attention has been given to the variousmechanisms that states could and should use to monitor quality in theirwaiver programs. As described in appendix VI, the initiatives CMS hasunder way include identification of strategies that states are currentlyusing to monitor and improve quality in home and community-based care,distribution of a guide on quality improvement and assessmentmechanisms for states and regional offices, and provision of a variety of technical assistance and resources to states. The agency also hasimplemented a new HCBS waiver quality review protocol for use byregional offices in assessing whether state waivers should be renewed. 29 Regional office staff told us that some states have begun to modify theirapproaches to quality assurance in HCBS waivers based on the use of thenew waiver review protocol. For example, Washington officialsestablished a new quality assurance unit within the agency that overseesits waiver for the elderly. In May 2002, CMS also introduced a voluntaryapplication template for its new consumer-directed HCBS waiver that asksfor a detailed description of states quality assurance and improvement

programs, including (1) the frequency of quality assurance activities,(2) the dimensions monitored, (3) the qualifications of quality assurance

staff, (4) the process for identifying problems, including sampling

29This protocol was developed to provide a standardized and comprehensive set of procedures for regional office staff to follow when conducting periodic waiver reviews. SeeDepartment of Health and Human Services, HCFA, HCFA Regional Office Protocol for Conducting Full Reviews of State Medicaid Home and Community-Based ServicesWaiver Programs (Washington, D.C.: Department of Health and Human Services, Dec. 20,2000).

CMS Lacks DetailedGuidance for States on theNecessary Components of a Quality AssuranceSystem


29/70


methodologies, (5) provisions for addressing problems in a timely manner,and (6) the system for handling critical incidents or events. While theseCMS activities are intended to facilitate the development of HCBS-relatedquality assurance approaches, they do not constitute a consistent set of minimum requirements and guidance for states use to obtain approval fortheir HCBS programs.

In addition to the lack of detailed guidance for states, CMS is not holdingits own regional offices or states accountable for oversight of the quality of care provided to individuals served under HCBS waivers. CMS regionaloffices are expected to conduct periodic waiver reviews to determinewhether states are protecting the health and welfare of waiverbeneficiaries. Annual state reports are required by statute, and CMSregulations indicate that they are intended to play a key role indetermining whether a waiver should be renewed. 30 We found that regionaloffices are neither conducting waiver reviews prior to renewal norobtaining complete annual state reports in a timely manner. As a result,CMS has not fully complied with the statutory and regulatory requirementsthat condition the renewal of HCBS waivers on states fulfilling theirassurances that necessary safeguards are in place to protect the health andwelfare of waiver beneficiaries.

Most CMS regional offices have not conducted timely reviews of the stateagencies administering waivers serving the elderly and other target

populations or completed reports to document the results of their reviews.Periodic on-site reviews are used to determine, among other things,whether a state is ensuring the health and welfare of waiver beneficiaries.Guidance from CMS headquarters instructs the regional offices to conductreviews before the first renewal of a waiver at the end of 3 years andwithin 5 years for subsequent waiver renewals.

Eighteen percent of all HCBS waivers (42 of 228) that have been in placefor 3 years or more as of June 2002 either have never been reviewed by the

regional offices or had not been reviewed prior to their last waiverrenewal. Approximately 132,000 beneficiaries were served by these 42waivers in 1999. Fourteen of the 42 waiversserving approximately 37,000waiver beneficiaries in 1999have had 10 or more years elapse without aregional office review (see table 5). CMSs Dallas regional office was

30 See, 50 Fed. Reg. 10013, 10016-17 (1985).

CMS Is Not HoldingRegional Offices or States

Accountable for Oversightof HCBS Waiver Quality

CMS Regional Offices Often Are Not Conducting TimelyReviews of State HCBS Waivers


30/70


responsible for 9 of these 14 waivers. Over a 10-year period, a regionaloffice should have conducted at least two reviews for each waiver. TheNew Mexico AIDS Waiver, initially approved in June 1987, has been in

place the longest without ever being reviewed15 years. CMS officialswere aware that regional offices had not reviewed some waivers but wereunaware of the extent of the problem.

Table 5: HCBS Waivers That Had 10 Years or More Elapse without Ever Having a Regional Office Review or without a ReviewPrior to the Last Waiver Renewal, as of June 2002

State Target populationNumber ofwaiver

beneficiaries a

Number of

years without aCMS regionaloffice review

No regional office waiver review ever conductedDallas regional officeNew Mexico Persons with AIDS 60 15Oklahoma Persons with mental retardation 2,550 b 14Texas Medically dependent children 895 b 14Louisiana Elderly and persons with disabilities 393 12New Mexico Medically fragile children 152 11Texas Persons with mental retardation and

related conditions1,047 11

Texas Persons with mental retardation 4,956 b 10Texas Persons with mental retardation 224 b 10Louisiana Elderly and persons with disabilities 113 10Seattle regional officeIdaho Elderly and persons with disabilities 1,000 12Idaho Persons with mental retardation and

developmental disabilities512 12

No regional office waiver review conducted prior to last waiver renewalKansas City regional officeIowa Elderly 3,994 11Missouri Elderly 20,821 10San Francisco regional officeHawaii Persons with AIDS 66 12

Source: CMS.

Note: GAO analysis of data provided by CMS, June 2002.aThe number of HCBS waiver beneficiaries is based on 1999 HCFA Form 372 data. See Harrington,Aug. 2001.bAuthors estimate. See Harrington, Aug. 2001.


31/70


As of June 2002, based on an analysis of the most recent regional officereview that occurred prior to October 2001 for each of the waivers, wefound that 23 percent of the review reports (36 of 158) in over half of theregional offices had not been finalized. 31 CMS requires its regional officesto prepare a final report on each HCBS review to document their findings,recommendations, and the state response. Without such a final report,there is no formal document to indicate whether a state has fulfilled therequired assurances, including those related to the health and welfare of waiver beneficiaries. The New York regional office did not finalize 11 of its12 reviews, dating back to 1998, and the San Francisco regional office didnot finalize 7 of its 13 reviews, 1 of which was for a review that occurred in1990. Without a final report documenting the review results, CMS cannotbe assured that, if problems were identified, they were appropriatelyaddressed.

Many state annual waiver reports submitted to CMS regional offices areneither timely nor complete. During the interval between regional officereviews, the required annual state waiver reports provide key informationon how states monitor beneficiaries quality of care and on any quality-of-care related problems. According to regional office officials, statesroutinely fail to submit these annual reports within the required timeframewithin 6 months after the period covered. In August 2000, officialsin CMSs Philadelphia regional office reported that they had currentannual state reports for less than half (11 of 28) of the waiver programs intheir region. Our review of the most recent annual state reports for 70 of 79 HCBS waivers serving the elderly confirmed that producing thesereports remains a problem: (1) reports for more than a third of the waiverswere at least 1 year latethe most recent report from one of LouisianasHCBS waivers was for calendar year 1997, (2) reports for approximatelyone-fourth of the waivers provided no information on whether deficiencieshad been identified through the monitoring processes, 32 and (3) fivereports indicated that deficiencies had been identified but provided no

31In our analysis, we included only those reviews that had taken place prior to October2001, allowing 9 months from the time the regional office conducted the waiver review tofinal report issuancefrom October 2001 to June 2002. CMS allows up to 4 months fromthe time the regional office completes all waiver review activities to issuance of a finalreport documenting the review findings.32 As noted earlier, about one-quarter of annual state reports for waivers serving the elderlydid not include information requested concerning the approaches used to monitor qualityassurance.

CMS Does Not Obtain Timelyand Complete State AnnualWaiver Reports


32/70


additional information about the nature of or response to the problems. 33 CMS headquarters has no central repository for annual state reports but isin the process of establishing a centralized database for state reportinformation sometime in 2003, a development that could facilitate ongoingmonitoring of the timeliness and completeness of these reports.

Our analysis of CMSs oversight activities for the 15 largest HCBS waiversserving the elderly demonstrates the extent of oversight weaknesses.Overall, 8 of the 10 CMS regional offices provided inadequate oversight for13 of these 15 largest state waivers for the elderly, which, in 1999, servedabout 215,000 beneficiariesover half (57 percent) of the total elderlywaiver beneficiary population at that time (see table 6). We found that

Four of the 15 HCBS waivers were not reviewed in a timely manner by theCMS regional officenone of the 4 had reviews for 8 or more years and

yet were renewed. 34 Four of the 15 waivers had no waiver review final report completed by the

regional office. Two of the reviews occurred in 1999, and for the remaining2 waivers the regional office could not tell us the date of the reviews orwhether a final report was available.

Four of the 15 waivers lacked a timely annual state report to the regionaloffice. As of April 2002, the most recent annual report for these 4 waiverswas either for the waiver period ending August 1999 (1 waiver) orSeptember 2000 (3 waivers).

Seven of the 15 waivers had annual state reports that were incompletebecause they either lacked information on their quality assurancemechanisms or on whether deficiencies had been identified.

33Eight of the remaining 9 waivers were new and had not yet had an annual reportsubmitted. The CMS Atlanta regional office did not provide a current annual report for 1waiver. As of June 2002, there were 77 waivers serving the elderly. However, our analysisincludes 2 additional waivers for the elderly that had been terminated or not renewed as of that date because the state was able to provide us with their most recent annual report.34These 4 waivers are a subset of the 42 HCBS waivers in place for 3 years or more thateither were never reviewed by the regional offices or were not reviewed prior to their lastrenewal.

Extent of OversightWeaknesses Evident in 15Largest Waivers Serving theElderly


33/70


Table 6: Status of CMS and State Monitoring for the 15 Largest HCBS Waivers Serving the Elderly

StateNumber of waiver

beneficiaries a

CMS waiver review nottimely or report not

finalized

Annual state report nottimely or documentation

insufficient b New York regional officeNew York 19,732 X XPhiladelphia regional officeVirginia 10,514 XAtlanta regional officeSouth Carolina 14,361 X c XGeorgia 14,018 XFlorida 13,762 XKentucky 13,339 XNorth Carolina 11,159 X c XChicago regional officeOhio 26,135Illinois 17,396 d X XWisconsin 13,900 XDallas regional officeTexas 27,978 X XKansas City regional officeMissouri 20,821 XDenver regional office

Colorado 11,481 XSeattle regional officeOregon 26,410 XWashington 25,718

Source: CMS.

Note: GAO analysis of data provided by CMS, June 2002 and the most recent annual state waiverreports. The 15 largest HCBS waivers serving the elderly are based on the number of beneficiaries.aThe number of HCBS waiver beneficiaries is based on 1999 HCFA Form 372 data. See Harrington,Aug. 2001.bThe annual report is required by statute and CMS directs states to (1) submit such reports within 6months after the period covered, and (2) include information on how the state implements, monitors,and enforces its health and welfare standards and the waivers impact on the health and welfare ofbeneficiaries.cThe CMS regional office could not provide the date that the last waiver review was conducted orspecify whether a report had been finalized.dAuthors estimate. See Harrington, Aug, 2001.


34/70


The limited scope and duration of periodic regional office waiver reviewsraise a question about the confidence that can be placed in findings aboutthe health and welfare of waiver beneficiaries. CMS regional officesconduct reviews using guidance provided by headquarters. The guidanceinstructs regional office staff to review beneficiary records; interviewwaiver beneficiaries, primary direct-care staff of waiver providers, andcase managers; and observe waiver beneficiaries and the interactionbetween the beneficiary and direct-care staff. This guidance was updatedin January 2001 when use of the new HCBS waiver quality review protocolbecame mandatory. However, the new protocol does not addressimportant operational issues such as

an adequate sample size or sampling methodology for the beneficiaryrecord reviews and interviews to provide a basis for generalizing thereview findings;

whether the sample should be stratified according to the different groupsserved under the waiver (i.e., for a waiver serving both the elderly and thedisabled, selecting a stratified sample based on the proportion of personsaged 65 and over and those aged 18 to 64 with disabilities); and

the appropriate duration of an on-site review, taking into consideration thenumber of sites and beneficiaries covered in the waiver.

Our analysis of regional office review reports for 21 HCBS waivers servingthe elderly found that the reviews varied considerably in the number of beneficiary records evaluated and their method of determining the sample,

potentially limiting their ability to generalize findings from the sample tothe universe of waiver beneficiaries. 35 Specifically, we found a wide rangeof sample sizes in 15 of the 21 regional office reviews that included suchinformation. The sample sizes for record reviews ranged from 14beneficiaries (of 73 served) in the Boston regional office review of the

Vermont waiver to 100 beneficiaries (of 24,000 served) in the Seattleregional office review of the Washington waiver. (See app. VII for asummary of the sample sizes in the regional office reviews.) Eleven of the15 CMS waiver review reports included information on the specificnumber of beneficiaries interviewed or observed during the review;however, we could not determine whether beneficiary interviews orobservations had been conducted in other waiver reviews. The method bywhich the beneficiary record review samples were selected varied, with

35We requested that regional offices provide us with final reports for HCBS waivers servingthe elderly issued from October 1998 to May 2002. Eight of the 21 reviews we analyzedwere completed after CMSs new HCBS waiver quality review protocol was implemented.

Scope and Duration of Regional Office WaiverReviews Are Limited


35/70


some regional offices using randomized sampling methods, some basingtheir sample on geographic location, and others reporting no method of sample selection.

For most of these same 15 waivers serving the elderly, we found that theregional staff typically spent 5 days conducting the waiver reviewregardless of the number of waiver beneficiary records sampled or theoverall size of the waiver. However, the Seattle regional office staff conducted only three reviews in the past 4 years, targeting its largestHCBS waivers. For example, the regional office has spent 3 to 4 weeks perwaiver for the on-site portion of the review and another week for stateagency interviews and review of documents. Generally, the number of beneficiary records reviewed and beneficiaries interviewed is dependenton (1) the number of days allocated to the waiver review by a regionaloffice and (2) the number of regional office staff members available.

The limited number of assigned staff and available clinical specialists,coupled with insufficient travel funds allocated to regional office oversightof HCBS waivers, have contributed to the timeliness and scope problemswe identified. According to regional offices, the level of attention given toHCBS waiver oversight, including periodic reviews when waivers come upfor renewal, is at the discretion of regional office management andcompetes with other workload priorities. 36 In August 2000, some regionaloffice officials formally communicated to HCFA headquarters theirconcern that the agency was not devoting sufficient resources to properlymonitor the quality of HCBS waiver programs. Regional office officialsresponsible for waiver oversight told us that the number of staff availablefor waiver oversight has not kept pace with the growth in the number of waivers and beneficiaries served and that resource issues remain a keychallenge for waiver oversight.

36Headquarters officials are responsible for establishing waiver policy and the 10 regionaloffices have responsibility for waiver oversight. Both headquarters and the regional officesanswer separately to the Administrator without any formal reporting links. In earlier work,we reported that these organizational reporting lines complicated coordination andcommunication, weakened oversight, and blurred accountability when problems arose. SeeU.S. General Accounting Office, Medicare Con tractors: Further I mprovement Needed in

Headquarters and Regional Office Oversight , GAO/HEHS-00-46 (Washington, D.C.: Mar.23, 2000) and U.S. General Accounting Office, Nursing Homes: Sustained Efforts Are

Essential to Realize Potential of the Quality Initiatives , GAO/HEHS-00-197 (Washington,D.C.: Sept. 28, 2000).

Limited Regional OfficeResources Available forOversight of HCBS Waivers
http://www.gao.gov/cgi-bin/getrpt?GAO/HEHS-00-46http://www.gao.gov/cgi-bin/getrpt?GAO/HEHS-00-46http://www.gao.gov/cgi-bin/getrpt?GAO/HEHS-00-46http://www.gao.gov/cgi-bin/getrpt?GAO/HEHS-00-46http://www.gao.gov/cgi-bin/getrpt?GAO/HEHS-00-197http://www.gao.gov/cgi-bin/getrpt?GAO/HEHS-00-197http://www.gao.gov/cgi-bin/getrpt?GAO/HEHS-00-197http://www.gao.gov/cgi-bin/getrpt?GAO/HEHS-00-197http://www.gao.gov/cgi-bin/getrpt?GAO/HEHS-00-46


36/70


We found that CMS regional offices differed substantially in the number of staff assigned to waiver oversight and the extent to which staff withclinical or program expertise were assigned to waiver oversight.

According to Dallas, Denver, and Philadelphia regional office staff, thelevel of resources allocated by the regional offices for such reviewsdictated the number of waiver beneficiary records reviewed or beneficiaryinterviews conducted. Six of the 10 regional offices had two or fewer full-time-equivalent (FTE) staff assigned to monitoring HCBS waivers (seetable 7). 37 Moreover, we found that the number of regional office staff assigned to monitoring HCBS waivers bore little relationship to the waiverworkload. For example, the Chicago regional office had six FTE staff tomonitor 34 HCBS waivers with 131,902

Gao Report Fed Oversite Hcbs Needs Strengthened

Documents

Transcript of Gao Report Fed Oversite Hcbs Needs Strengthened